This project will determine what components of the immune response are involved in the in vivo human tumor host relationship. The immune cell infiltrate in and around lymphoid neoplasms will be studied in frozen tissue sections. Different types of T cells, B cells, and accessory cells will be identified with various hybridoma monoclonal antibodies and detection methods employing antibodies, biotin, avidin, and horseradish peroxidase. Methodology for enumerating cells in sections will allow accurate comparison of cell types in multiple neoplasms of similar type as well as with those in normal and hyperplastic lymphoid tissues. The type of immune infiltrate will be correlated with tumor phenotype and expression of HLA antigens. The type and amount of infiltrate will also be compared with anatomic site of involvement and various clinical paramters in an attempt to predict certain clinical characteristics. The categorizing of immune infiltrates may prove useful in experimental treatment strategies employing monoclonal antibodies coupled to particular effector cells. A long-range goal is to assess the stability of tumor phenotype and immune infiltrate at similar and differing sites as a function of time and treatment. New double-label methods and fixation and processing techniques will be developed and refined to allow extremely precise localization of cellular antigens. Such an in vivo evaluation of tumor host relationship should provide valuable insights in tumor biology. (IP)